So it is not about efficiency anymore but about raw power?
How can you talk about efficiency without involving raw power?
It makes no sense.
You can demolish an abandoned village faster with a nuke than with construction vehicles operated by people but a nuke isn’t very efficient despite being effective.
My sports car will drive faster laps than a normal car but the fuel economy is bad, even when they would drive the same speed. It’s not very efficient despite having lots of raw power.
For efficiency we normally talk about not wasting any power.
My point was simply that on my i7-4900MQ test system, Krita performs really good despite being an old, hot mobile workstation platform. That doesn’t prove krita is perfectly optimized, but it does suggest that your results may be influenced by factors specific to your system rather than proving a general optimization problem in krita.
“Most of the time, it’s just because of my PC’s thermal paste wearing off.”
That weakens your earlier argument considerably because it introduces a major hardware variable affecting the temperatures you originally relied on.
And you’re substituting “new hardware is faster” for “krita is poorly optimized”
The first statement is true. arguing that newer hardware is faster. Nobody disagrees with that. The second still hasn’t been demonstrated.
also krita is written in C++ Krita - Wikipedia
so I’m not sure why bytecode is relevant here. Newer hardware executing code faster is expected, but that doesn’t tell us whether Krita itself is efficient or inefficient.The burden of proof doesn’t shift to everyone else just because a theory sounds plausible. If youre claiming krita is inefficient, it’s your responsibility to demonstrate that. And i think asking others to disprove it while using terms like “bytecode” incorrectly doesn’t strengthen the argument
The optimization problems of Krita are honestly relative.
If you do not compare to how other apps perform for the same specific task,
it’s almost useless to speak of optimization with Krita. Issues would arise
only when an operation is way too slow.
For example, the liquify tool in Krita, is reputed for having insufficient
speed/performance so far, especially when the layer data is big. That
is a practical case of poor optimization because in CSP or Photoshop,
the liquify tool is super fast, even with high pixel density on a layer.
If you do not use measurement metrics or compare to other software
for a specific task, or a tool that achieve the same result between the softwares
of comparison, how are you gonna speak of optimization?
My hardware is also a 3rd gen core i7. Hot like yours too when
I use Krita. But when I said that Krita consumes tons of energy,
it is relative. Softwares like Photoshop, Clip Studio and Leonardo, in
some instances, demand less energy to achieve the same specific task.
Ultimately, the Krita of today is much better the Krita of many years ago.
Even if there’s still a substantial mile to go again, it has known many optimizations
over time, and is more bearable for production. With time it shall improve even further
whenever possible. I’m not complaining right now as there is not much I can do
about the software’s development pace.